Fecal Microbiota Enterotypes of Preterm Infants at the Neonatal Intensive Care Unit (NICU) in Association with Dietary and Clinical Factors

Abstract Objectives The gut microbiota of preterm infants (PTI) differs from that of term infants, with higher abundances of pathogenic bacteria and late acquisition of beneficial bacteria. This dysbiosis is affected by different types of milk and milk fortifiers fed to PTI, exposure to antibiotics after birth, and long hospitalization periods. Different enterotypes have been proposed to classify the gut bacteria ecosystems in adults, but little data exits regarding the PTI gut microbiota. Thus, the objective herein was to investigate gut microbial enterotypes of PTI infants. Methods PTI were followed from birth until NICU discharge. Data including daily feeding information and medications were obtained from the medical records. Freshly voided stool samples were collected, bacterial DNA was extracted and the V3-V4 regions of the 16S rRNA were sequenced. Enterotypes were determined using the partitioning around medoids clustering algorithm and the Jensen-Shannon divergence method using RStudio. Results A total of 551 stool samples were collected from 97 PTI. At genus level, two enterotypes were obtained; enterotype A (EA) was characterized by a high abundance (62%) of Escherichia-Shigella and Staphylococcus, whereas Enterobacteriaceae, Clostridium sensu stricto 1 and Klebsiella accounted 55% of relative abundance for Enterotype B (EB). Alpha diversity (Shannon index) was higher (P < 0.0001) in EB. In the earliest sample collected after birth (2.2 ± 1.1 weeks of life), the majority of PTI (64%) belonged to EB, but 37% of PTI switched enterotypes during their hospital stay, most of these changed from EA to EB. The change on enterotypes occurred at 4.6 ± 2.7 weeks of life. Bovine milk-based fortifier (BMF) and abundance of Escherichia-Shigella were positively associated in EA, whereas, this correlation was negative for EB. Similarly, Enterobacteriaceae abundance was positively correlated with the use of antibiotics in EA, but was negatively correlated in EB. Conclusions The gut microbiota of PTI was more likely to belong to a more diverse enterotype. There were opposite effects between both enterotypes to exposure to BMF and antibiotics. This suggests that responses to dietary and clinical factors could be dependent upon the characteristics of the gut microbiota of PTI. Funding Sources Seed grant Carle Foundation Hospital and University of Illinois. CONACyT Graduate Fellowship.